
HKU launches investigation after paper containing non-existent references generated by AI prompts student misconduct probe
The University of Hong Kong (HKU) has initiated an internal investigation after a published academic paper, led by a PhD candidate, was found to contain multiple non-existent references apparently generated by artificial intelligence (AI).
The corresponding author, Professor Paul Yip Siu-fai of HKU’s Department of Social Work and Social Administration, issued an apology on Sunday on behalf of himself and the lead author, Ms Bai Yiming.
The issue was raised when a user on the social-media platform Threads flagged the paper’s reference list as potentially the output of “AI hallucination”—a term used to describe instances where AI systems fabricate plausible-looking but non-existent information.
Professor Yip acknowledged that Ms Bai had used AI to assist with the referencing process without individually verifying each citation, and he accepted oversight responsibility as corresponding author.
In his statement, Professor Yip asserted that while the agreement to submit the paper followed two rounds of peer review, he believed the paper’s findings were not themselves fabricated and therefore did not constitute academic misconduct in his view.
He noted that other co-authors served only in support roles and did not contribute to the reference validation process.
The university said that it had established stringent policies on AI use and had asked the relevant faculty to conduct a full assessment of the matter.
The case reflects wider challenges in higher education concerning the responsible integration of generative AI in academic workflows.
Universities globally are adapting policy frameworks to address risks such as reference fabrication, ghost-writing and other integrity concerns, and HKU’s investigation may yield precedents for emerging oversight mechanisms.
As the probe unfolds, attention will focus on the university’s governance response, the extent of policy enforcement and whether academic sanctions will follow.
HKU declined to comment on the disciplinary outcome but affirmed its commitment to upholding academic integrity and reviewing procedures for the use of AI tools in research.
The investigation is ongoing.
The corresponding author, Professor Paul Yip Siu-fai of HKU’s Department of Social Work and Social Administration, issued an apology on Sunday on behalf of himself and the lead author, Ms Bai Yiming.
The issue was raised when a user on the social-media platform Threads flagged the paper’s reference list as potentially the output of “AI hallucination”—a term used to describe instances where AI systems fabricate plausible-looking but non-existent information.
Professor Yip acknowledged that Ms Bai had used AI to assist with the referencing process without individually verifying each citation, and he accepted oversight responsibility as corresponding author.
In his statement, Professor Yip asserted that while the agreement to submit the paper followed two rounds of peer review, he believed the paper’s findings were not themselves fabricated and therefore did not constitute academic misconduct in his view.
He noted that other co-authors served only in support roles and did not contribute to the reference validation process.
The university said that it had established stringent policies on AI use and had asked the relevant faculty to conduct a full assessment of the matter.
The case reflects wider challenges in higher education concerning the responsible integration of generative AI in academic workflows.
Universities globally are adapting policy frameworks to address risks such as reference fabrication, ghost-writing and other integrity concerns, and HKU’s investigation may yield precedents for emerging oversight mechanisms.
As the probe unfolds, attention will focus on the university’s governance response, the extent of policy enforcement and whether academic sanctions will follow.
HKU declined to comment on the disciplinary outcome but affirmed its commitment to upholding academic integrity and reviewing procedures for the use of AI tools in research.
The investigation is ongoing.







































